Numerical Aerodynamic Simulation Program Long Haul Communications Prototype

144 Pages·2004·5.64 MB·English

https://ntrs.nasa.gov/search.jsp?R=19870008921 2019-04-05T13:36:42+00:00Z
NAKTeckni ca I Memorandum 89406
Numerical Aerodynamic
Simulation Program Long Haul
Communications Prototype
Bohden K. Cmaylo and Foo Lee
( L AS A -IB-894O 6) B UPlE ti IC AL A E E CDY h AaIC N87- 18354
SILIDLATICN PSCGBAk LCLG HAUL CCBEUNICA5IUNS
PhC'IO'IYPE (NASA) 144 p CSCL 178
UnclaE
G3/62 43524
January 1987
c
National Aeronautics and
Space Administration
NASA Technical Memorandum 89406
4
Numerical Aerodynamic
Simulation Program Long Haul
Communications Prototype
Bohden K. Cmaylo, Ames Research Center, Moffett Field, California
Foo Lee, General Electric Company, San Jose, California
January 1987
?
National Aeronautics and
Space Administration
Ames Research Center
Moffett Field, California 94035
TABLE OF CONTENTS
Page
............................................................................
SUMMARY i
. ...................................................................
1 INTRODUCTION 1
..................................................................
1.1 Purpose 2
....................................................................
1.2 Scope 2
. ...........................................................
2 APPLICABLE DOCUMENTS 3
. ..............................................................
3 EXECUTIVE SUMMARY 4
.............................
3.1 Initial Objectives and Results of Objectives 4
..........................................................
3.2 Accomplishments 6
..........................................................
3.3 Recommendations 8
. ...............................................................
4 TECHNICAL REPORT 9
...............................................
4.1 LHCP Configuration Changes 9
........................................................
4.2 LHCP Experiments 11
...................................................
4.3 Resource Requirements 14
.........................................................
APPENDIX A ABBREVIATIONS 23
.............................
APPENDIX B LHCP COMMUNICATIONS AND TESTING EQUIPMENT 28
......................................................
APPENDIX C LHCP EXPERIMENTS 37
..........................................
APPENDIX D LHCP REMOTE MONTHLY REPORTS 104
...........................................
APPENDIX E PROBLEM HISTORY/RESOLUTION 125
........
APPENDIX F CONTRIBUTORS TO THE LONG HAUL COMMUNICATIONS PROTOTYPE REPORT 140
i ii
NUMERICAL AERODYNAMIC SIMULATION PROGRAM LONG HAUL COMMUNICATIONS PROTOTYPE
Bohden K. Cmaylo and Foo Lee*
Ames Research Center
SUMMARY
This document is a report of the Numerical Aerodynamic Simulation (NAS) Long
Haul Communications Prototype (LHCP). It describes the accomplishments of the LHCP
group, presents the results from all LHCP experiments and testing activities, makes
recommendations for present and future LHC activities, and evaluates the remote
workstation accesses from Langley Research Center, Lewis Research Center, and
Colorado State University to Ames Research Center. The report is the final effort
of the Long Haul (Wideband) Communications Prototype Plan (PT-1133-02-N00),
3 October 1985, which defined the requirements for the development, test, and opera-
tion of the LHCP network and was the plan used to evaluate the remote user bandwidth
requirements for the Numerical Aerodynamic Simulation Processing System Network.
1. INTRODUCTION
This report consists of four sections and six appendixes. A short description
of each follows:
Section 1: An overview of the purpose and scope of the Long Haul (Wideband)
Communications Prototype (LHCP) network.
Section 2: Referenced documents used in the preparation of this report.
Section 3: Summary of the objectives of the LHCP and the results of each
objective, and the accomplishments and recommendations of the LHCP group.
Section 4: The technical report, which describes in detail how the objectives
were accomplished; this section discusses the LHCP configuration, experiments, and
resource requirements.
Appendix A: List of acronyms used in this report.
Appendix B: Description of the LHCP hardware and Electronics Laboratory test-
ing equipment.
Appendix C: Reports of the LHCP experiments.
"General Electric Company, San Jose, California.
1
Appendix D: Monthly reports received from the remote sites by the LHCP group.
Appendix E: Problem history and resolution during the course of establishing
the LHCP.
Appendix F: Contributors to the Long Haul Communications Prototype Report.
1.1 Purpose
.
The purpose of this document is to report the steps taken in the development,
test, and operation of the Long Haul (Wideband) Communications Prototype (LHCP).
(See Numerical Aerodynamic Simulation (NAS) Projects Office ref. 1 for the LHCP
Plan, which proposed the initial design and direction of the LHCP network. Refer-
ences are listed in Section 2, Applicable Documents, of this paper.) This LHCP
network was designed to evaluate the remote user bandwidth requirements for the NAS
Processing System Network (NPSN), and to initiate access to the NPSN Initial Oper-
ating Configuration (IOC) by remote computational fluid dynamics users. (See NAS
Projects Office ref. 2 for a more detailed description of the IOC.) The NPSN is
being developed under the direction of the NAS Projects Office of the National Aero-
nautics and Space Administration (NASA), Ames Research Center, Moffett Field,
California.
This report defines and evaluates the remote workstation accesses to Ames
Research Center (ARC) from the NASA Langley Research Center (LaRC), Hampton,
Virginia; the NASA Lewis Research Center (LeRC), Cleveland, Ohio; and Colorado State
University (CSU), Fort Collins, Colorado.
1.2 Scope
The LHCP effort was used to establish computer communications between ARC and
the remote sites, and to evaluate the Vitalink VB/1 bridge boxes, TransLAN software,
communications equipment, communications links, and the bandwidth requirements of
the remote users.
The LHCP effort consisted of using NAS workstations at LaRC and LeRC, and a
non-NAS workstation at CSU, all of which are connected to their local area networks
via Ethernets, to access the NPSN via a gateway host at ARC.
Three experiments--the user bandwidth experiment, the hardware experiment, and
the transmission control protocol/internet protocol (TCP/IP) tuning experiment--were
used to evaluate the LHCP. The report on the user bandwidth experiment can be found
in NAS Projects Office reference 8, and the reports on the hardware and TCP/IP
experiments are presented in appendix C.
The results of the experiments were used to establish the remote user needs
within the development of the LHC subsystem. The LHCP provided NAS personnel with
valuable experience in wideband communications and computer networking. This
experience aided in the design, development, implementation, and operation of the
LHC subsystem portion of the Integrated Support Processor Complex (ISPC). (See NAS
Projects Office ref. 4 for a more detailed description of the ISPC.)
The LHCP effort (i.e., the prototype as a whole) was divided into two phases,
which allowed early user involvement in the development of the LHCP network (i.e.,
the wideband communications links).
1. The first phase used remote NAS workstations (Silicon Graphics, Inc.'s
(SGI) IRIS 1500s, which were upgraded to 2500s), a remote non-NAS workstation
(Sun 160 workstation), and the NAS Ethernet network via a VAX 11/780 (the Cray
gateway) to access the Cray X-MP/12. This phase allowed LHCP testing to start
before the NAS Cray-2 was connected to the NPSN. (See NAS Projects Office ref. 3
for a description of the IRIS workstation, and NAS Projects Office ref. 2 for a
description of the Cray.)
2. The second phase used both NAS and non-NAS type remote workstations to
access the Cray-2 when it was operational and connected to the NPSN.
These phases did not include additional bandwidth capabilities, such as T1 data
rate (1.544 megabits per second, Mbps), since that capability was not available for
use at that time. The LHCP high-speed wideband TI network will be supplied by the
NASA Program Support Communications Network (PSCN) when it becomes available. The
PSCN contract was awarded to Boeing Computer Services on 1 April 1985. A commercial
communications network was used in the early phases of this experiment. Boeing/RCA
supplied the interim service at 224 kbps; the 56-kbps line was obtained from
Boeing/AT&T. All NASA communications requests for resources, such as NASA PSCN
circuits and interim circuits, were channeled through the Marshall Space Flight
Center. The circuits were transferred to PSCN administration and control on 1 April
1986.
2. APPLICABLE DOCUMENTS
Although the LHCP plan is not a controlled document, there was an effort to
make it consistent with the other documents governing the NAS program. The applica-
ble NAS documents are listed below.
1. PT-1133-02-NOO Numerical Aerodynamic Simulation Program
Long Haul (Wideband) Communications Prototype Plan for Remote
Workstation Access to ARC from LaRC, LeRC, and CSU.
3 October 1985
2. PP-1100-01-CO1 Numerical Aerodynamic Simulation Program
Development Plan for the NPSN Initial Operating Configuration
24 April 1985
3
3. ~~-1126-00-NOO Numerical Aerodynamic Simulation Program
User Interface Prototype Development Plan
14 May 1984
4. RFP2-31373(DHC) Integrated Support Processing Complex (ISPC)
Numerical Aerodynamic Simulation
Processing System Network (NPSN)
5. PT-1133-01-NOO Numerical Aerodynamic Simulation Program
Long Haul Communications Prototype Experiments Plan
24 June 1985
6. TBD Numerical Aerodynamic Simulation Program
Operational Policy and Procedures for Remote Workstations
26 February 1986
Program Configuration Control Board Recommended 25 March 1986
7. TBD Numerical Aerodynamic Simulation Program
Long Haul Communications Prototype Transition to Operations
Plan
21 April 1986
8. TBD Numerical Aerodynamic Simulation Program
Long Haul Communications Prototype Experiments Results
(to be determined)
3. EXECUTIVE SUMMARY
This summary consists of three subsections. The first subsection sets forth
the initial objectives of the LHCP and discusses how each objective was satisfied.
The second subsection reports the accomplishments of the LHCP group in establishing
the network. The third subsection presents the recommendations of the LHCP group as
they relate to future NAS Projects Office activities dealing with long-haul
communications.
3.1 Initial Objectives and Results of Objectives
The results of the initial objectives, as defined in NAS Projects Office
reference 1, are summarized below after each of the stated initial objectives. The
initial objectives were as follows.
1. Establish and evaluate the LHCP wideband network design. The LHCP wideband
network design that was established consisted of three remote sites: LaRC, LeRC,
and CSU. Langley Research Center and LeRC are communicating with ARC over 224-kbps
satellite links provided by BoeingIRCA. Colorado State University is communicating
4
with ARC over a 56-kbps terrestrial link provided by Boeing/AT&T. Vitalink VB/1
bridges and related TransLAN software are used to connect the remote sites with
ARC. The evaluations of the Vitalink VB/1 bridges have shown that their use is a
feasible method of meeting long-haul communications requirements and providing long-
haul communications capabilities to remote sites at those data transmission rates.
The LHCP hardware experiment consisted of evaluating the Vitalink VB/1 bridge boxes,
learning how to use the TransLAN software, the satellite delay simulator, modems,
and other communications equipment; and evaluating the communications links. Sev-
eral hardware tests were designed to determine workstation data transmission charac-
teristics and to measure data transfer rates of various sizes between worksta-
tions. One test evaluated transmissions between workstations that were on the same
Ethernet. Other tests evaluated transmissions between workstations connected to two
separate Ethernets but bridged together with VB/ls over a simulated communications
link and over the 56- and 224-kbps communications links. The complete report on the
hardware experiments is presented in appendix C.
2. Initiate early remote user activity. The LHCP effort initiated early
remote-user activities by allowing the remote users from LaRC, LeRC, and CSU to
access NPSN resources before IOC and the establishment of the LHC subsystem.
3. Evaluate and establish the LHCP bandwidth requirements. The LHCP bandwidth
experiment as defined in NAS Projects Office reference 5 started in the third quar-
ter of 1985, before the NPSN IOC was fully operational, and continued over several
months. The purpose of this experiment was to provide experience in high-speed
communications, initially at 56 and 224 kbps, and, it was hoped, at 1.544 Mbps
(which did not occur), and to provide experience with remote communications
capabilities. Emphasis was on the evaluation of remote NAS workstation
communications from both LadC and LeRC to ARC and remote non-NAS workstation
communications from CSU to ARC. The results from the bandwidth experiment will be
used to help establish the bandwidth requirements for future remote sites wishing to
get on the LHC subsystem. (See NAS Projects Office ref. 5 for a detailed
description of the LHCP user bandwidth experiment plan.)
4. Generate cooperative user involvement. The LHCP effort was also a means by
which to generate cooperative user involvement. The remote users (specifically the
site/system administrators) were involved from the very beginning of the implementa-
tion of the LHCP wideband communications network and installation of the Vitalink
and related communications equipment. The LHCP could not have been implemented
successfully without the cooperative involvement of the administrators.
5. Provide user feedback through use of NAS workstations (IRIS ISOO/ZSOO) from
remote-user sites. NAS workstations (IRIS 1500/2500) were shipped to LaRC and LeRC
for use from these remote sites to access the NPSN resources. Feedback regarding
the use of the NAS workstations and access to the NPSN resources was provided by the
remote users in form of monthly reports and trouble calls to the LHCP group. The
reports and trouble calls will be handled by the NAS Projects Office User Services
in the future.
5
6. Provide user feedback through use of a non-NAS workstation (Sun 160) from a
remoteusersite. A non-NAS workstation (Sun 160) was used from CSU to access the
NPSN resources. Feedbacks regarding the use of the non-NAS workstation and access
to the NPSN resources was provided by the remote users in form of monthly reports
and trouble calls to the LHCP group. The reports and trouble calls will continue to
be handled by the NAS Projects Office User Services.
7. Transfer knowledge and experience gained in working with PSCN and commer-
cial communications services to the IOC. The knowledge and experience gained in
working with the NASA PSCN, and commercial communications services have been trans-
ferred to the LHC subsystem group. The LHC subsystem group has used and will con-
tinue to use the knowledge to further develop the LHC subsystem for IOC implementa-
tion. (A list of problems and experiences encountered during the LHCP is included
in appendix E.)
8. Transfer development resources (i.e., hardware and software) acquired to
the10C. The functional Vitalink VB/1 bridges communicating with LaRC, LeRC, and
CSU were transferred to NAS Projects Office Operations. (See NAS Projects Office
ref. 7 for a full description of the transition to operational activities.) The
user services activities were transferred to NAS Projects Office User Services.
Additional development resources (i.e., hardware and software) acquired during the
LHCP including the LHC Electronics Laboratory, laboratory equipment, and
communications equipment will be transferred to the LHC subsystem group or to the
NAS Projects Office Operations or to both.
3.2 Accomplishments
The LHCP was concluded on 30 April 1986; the following had been accomplished.
1. The LHCP development plan was written and approved on 3 October 1985 by the
NAS Projects Office Engineering Review Board. The plan described the scope, activi-
ties, and schedules of the prototype. (NAS Projects Office ref. 1.)
2. This report on the LHCP development was written at the conclusion of LHCP
activities. It noted modifications to the original plan to better document the
actual activities of the LHCP. An executive summary, which included the accomplish-
ments and recommendations of the LHCP group, was also included with the technical
report. Appendixes were included that present the communication and testing hard-
ware descriptions, LHCP experiment reports, remote monthly reports, and the LHCP
problem history and resolution reports.
3. The Vitalink boxes (VB/l and CS/lOO) connected the remote sites to the
NPSN. The Vitalink communication equipment interfaced to the LHCP communication
circuits and joined the remote site Ethernets to the local NAS Processing System
Ethernet subnetwork. A checkout list of equipment was designed to facilitate con-
nections to NPSN from other remote sites.
6
4. Three remote sites were requested to participate in evaluating the LHCP:
CSU, LaRC, and LeRC. The following describes the activities establishing the LHCP
communication links with those sites. (a) A communication link was established via
a 56-kbps AT&T circuit from ARC to CSU. The Vitalink box (VB/1) was first tested by
the LHCP group, and then delivered to CSU on 22 July 1985. The circuit was acti-
vated on 21 June 1985. The connection of the Sun workstation at CSU to the NPSN at
ARC communication was established 2 August 1985. The long delay between circuit
activation and data communication was due to a delay in delivery of the Data Service
Unit (DSU) modems by Boeing. (b) A communication link was established via a
224-kbps RCA Americom Satellite circuit from ARC to LaRC. The VB/1 was delivered to
LaRC on 6 September 1985. The IRIS workstation at LaRC was on loan from the NAS
Projects Office for this purpose and was received by LaRC on 6 September 1985. The
satellite circuit was activated on 6 September 1985, and completely tested and
released by RCA on 30 September 1985. The connection of the IRIS workstation at
LaRC to the NPSN at ARC communication was established 10 October 1985. An upgrade
from the IRIS 1500 model to an IRIS 2500 model was completed by LaRC on 5 February
1986. (c) A communication link was established via a 224-kbps RCA Americom
Satellite circuit from ARC to LeRC. The VB/1 was delivered to LeRC on 12 September
1985. The IRIS workstation at LeRC was on loan from the NAS Projects Office for
this purpose and was received by LeRC on 12 September 1985. The satellite circuit
was activated on 16 September 1985, and completely tested and released by FiCA on
18 October 1985. The connection of the IRIS workstation at LeRC to the NPSN Network
at ARC communication was established 18 October 1985. An upgrade from the IRIS 1500
model to an IRIS 2500 model was completed by LeRC on 2 April 1986.
5. The LHCP experiments plan was written and approved on 6 June 1985 by the
NAS Projects Office Engineering Review Board. This plan described the user band-
width experiments scope, schedule, and activities that were to take place between
ARC, CSU, LaRC, and LeRC. This experiment was to evaluate user satisfaction with
the LHCP remote circuits. A primary result of this experiment was to obtain infor-
mation so that bandwidth requirements for future remote users could be evaluated
(NAS Projects Office ref. 5).
6. The user bandwidth experiment was partially completed. The experiment was
successful for the CSU participants; however, in performing the LaRC user bandwidth
experiment and the hardware experiment, it was discovered that the TCP/IP protocol
used did not perform satisfactorily over satellite links, and that the maximum
performance observed was approximately 10% of the 224-kbps satellite bandwidth.
Further experiments were canceled until either the problem is resolved or additional
bandwidth requirements data are needed (NAS Projects Office ref. 8).
7. The hardware experiment was designed to test various portions of the com-
munication circuits of the LHCP. These tests primarily tested the Ethernet transfer
rates to obtain a baseline, and Vitalink-to-Vitalink transfer rates with comparisons
to the baseline. The baseline portion was completed; however, the full Vitalink-to-
Vitalink tests are still in process. The results are presented in appendix C.
7